I did read the post. I'm suggesting that the OP doesn't understand statistics, independent events, and random outcomes. A truly random 20% chance at an outcome when it comes to independent events means that over a very large sample size (IE, not people complaining on forums, in gen chat, or from a few guilds), crafters will learn a schematic 20% of the time they attempt a RE. As I said before, this means you have an 80% chance to fail each time. The game doesn't care how many times you tried before.
An event can be random and produce what looks like a pattern, it can randomly produce outcomes that are bad for you, or it can randomly produce outcomes that are good for you. People love to complain about getting really lousy outcomes when it comes to RNG, but nobody posts with a complaint saying, "Bioware, I've been getting schematics on the first try too much, please nerf my RE rates."
A bit unrelated, but I'm betting it's actually pretty close to 20%, and the reason the RE rate will be dropped to 10% for the 72 tier of raid gear is that Bioware is unhappy with how quickly people geared out in full 63s.
Nobody has proved that the RE success are independent events. Nor has anyone proved that the chance is 20%. What the OP said is that IF the RE success rate are independent events that are evenly distributed, then he has a high confidence interval that it's not 20%.
The key words are 'Independent events', 'Evenly distributed', and '20%'. The first 2 has never been proven at all.